cylinder 2 & 4 rich/missfire
 

So, a week ago, I brought my car to have it tuned, and the tuner thought it was running a bit rough at idle. After pulling a couple of plugs, we found out that cylinders 2 & 4 were black and sooty. I'm sure this wasn't the case with the stock ecu before the turbo and megasquirt. :)

Brought the car home and did a bunch of stuff to try and figure it out. To me, it was narrowing down whether it was a problem with fuel or spark. For some of the swaps, I was trying to see whether the problem moved with the swap.

After each change, I'd go around the block or two and pull the plug. I'd clean it with a wire brush just so I could definitely could tell whether there was a change or not.


Spark:
moved the plugs around in different order
swapped back to original plugs NGK BKR5E11, from the BKR7E11
swapped the plugs wire on each individual coil bank
(My 94 is 3 & 2 on one bank, and 1 & 4 on the other bank)
swapped the coil pack to a different used coil pack
plug wires are fairly new, I could pull each plug and hear the rpm drop across each cylinder
when I do pull the plug wire and have it arc to ground, it does just about kill the motor. (the dyno guy said that was unsusual, is it?).

Fuel:
swapped injectors 2-->1, 1-->2, 4-->3, 3-->4
verified there was no vacuum leaks by spraying with carb cleaner
pulled injectors and verified o-rings were ok

Megasquirt:
Changed back to baseline tune

Other:
compression consistent across all 4 cylinders

So none of the above tests were fruitful. I'm thinking at this point to start checking out the ignitors at the Megasquirt. Maybe one of the chips has noise or not firing consistently...


Suggestions welcome.

You sure did a lot of testing.

Why, if you suspect the MS, did you not swap it out for the stock ECU?

Hmm, I guess swapping back the stock ecu would be a possibility provided I swap back to the stock injectors and MAF sensor. Sounds like too much trouble for the MAF, unless I could run with without it while testing just around the neighborhood.

As long as you have the maf unplugged, the car should just go into limp mode and be rich. Should start up,rev,idle fine without it though.

Doesn't really help with his plug reading then.

It may idle rich, but if it idles smooth and eliminates the "roughness" he and his tuner were experiencing, to me that would indicate an issue with his megasquirt (whether tune-wise or hardware). Of course the same test with the MAF attached is still the more ideal route, but.

So, so thats kind of a option then... swapping back to stock ECU..

On another note, because I have a '94 I have individual wires for each injector, as compared to 90-93 which has 1 wire for 2 & 4, and another for 1 & 3. My megaquirt is a MS1, so batch fire only, and not sequential injection like I originally thought.

Looking at the wiring diagrams for the DB37, 32 and 33 are connected together, and 34 and 35 are connected together. I didn't do this as I initially didn't understand it, and ended up running each injector wire to each pin.

Measuring the resistance between 32-33, and 34-35 at the megasquirt, it seems they are connected internally.

I guess I'll do the branching on the harness side as a test. If that doesn't get me anywhere, will do some resistance measurements to the injectors to see if there's any difference.

Ref:
https://www.miataturbo.net/attachmen...ine=1405798625

My diagram: (32 & 33, and 34 & 35 not connected together) Disregards grounds, they are grounded to block now.

https://www.miataturbo.net/members/b...ected-3581.jpg

You can use one pin for one injector; the two are, as you noted, bridged internally. Four injectors, four pins - you don't have to merge the injector wires externally.

Yeah, that was a long shot, thanks for clarifying that Matt. My next test will be to swap back the stock ecu and see what happens after that. Waiting for some stock injectors to arrive and will have to swap out my 650s for the stock ecu.

I can definitely tell its cutting out between 1500-1800rpm.

Do you suggest any other tests within the megasquirt to give it a clean bill of health?

Thanks

Here's the vid of the missfire. you can see from 1500-2000 how the idle drops

https://www.dropbox.com/s/ci5hmjwi5ztc3eb/missfire.MOV

Can you post a data log of the misfire and your MSQ? This will give more detailed information on what's going on.

Geez, missfire is not happening now where it shows on the tach, but plugs are still black, so pretty sure its still happening. Have included the logs
Thanks for checking them out!

Swapped back to my stock injectors and was about to swap back to the stock ecu, but forgot that I broke my narrowband o2 sensor when i pulled it out. Ended up leaving the stock injectors at the moment, with adjust required fuel settings.

Stupid hydraulic lifters starting making some noise. They have been acting up a bit lately and now they are full time tapping away. Changed the oil (synthetic 5w/20 which has served me well for quite some time), and no difference.

The oil was only a few months old, but very black. Figured with all the richness, the lifters must be very dirty at this point. Anyway, off to fix that now, dammit!

Placing my order for stock o2 sensor today

The ECU seems to have a good RPM signal, but is running pig rich.

I noticed you have an MS1 - do you have the capacitor mod on the second trigger input?

Well I did this, not sure whether it covers second trigger input though

-----
-How to mod a MegaSquirt-I PCB3.0 ECU-

How to MegaSquirt your Mazda Miata

"The new "cap mod" turns this into a low pass filter that helps avoid misfires caused by a noisy CMP signal. Solder a length of wire to a 0.1 uF capacitor, and connect this capacitor-on-a-wire between JS8 and SG or a ground hole in the proto area.

So drove the car around tonight and I was lucky the lifter clatter went away on it's own after about 20 minutes of driving!!!

Car ran really well on the stock injectors, could feel it was much more smooth and torquey on the low end drive around town.

Pulled the plugs and no missfire on 2 & 4!!! Plugs were beatifully normal... How can this be??? Well, at least its not the megasquirt, wiring, ignition.!!

This all points back to the ID 650 injectors and pigtails. Even when swapping injectors around earlier, cylinders 2 & 4 were still rich... so don't think i have 2 defective injectors. Pigtails are all fine as they all measure low impedance.

Only thing I can think is somehow the voltage is not consistent when using the pigtails or the fuel flow is somehow not right when using these injectors with the stock fuel rail and Walbro 190 fuel pump.

Hmmm, making progress at least :)

That's the mod I was referring to.

650 cc/min injectors are close to the limits of what an MS1 can do satisfactorily at idle.

Oh, never thought about injector limit size. So the symptoms make sense? Whats the safe max size for the MS1? 470s/550s should meet my fueling needs.

Picked up some red top 470's, will see how that works. Anyone else using larger injectors with their MS1?

Picked up some 470s allright, and yet another fun experience! Can't say I'm not learning tons from this project! sheesh!

https://www.miataturbo.net/engine-pe...t-crank-80396/